ArvindSelvaraj
commited on
Commit
•
0df70ac
1
Parent(s):
6035560
Update backend.py
Browse files- backend.py +65 -104
backend.py
CHANGED
@@ -1,12 +1,10 @@
|
|
1 |
import json
|
|
|
2 |
import io
|
3 |
import requests
|
4 |
import html # For escaping HTML characters
|
5 |
from bs4 import BeautifulSoup
|
6 |
-
import
|
7 |
-
from openpyxl import Workbook
|
8 |
-
from openpyxl.styles import Alignment, Font
|
9 |
-
from openai import OpenAI
|
10 |
|
11 |
# Initialize OpenAI API with Nvidia's Llama model
|
12 |
client = OpenAI(
|
@@ -26,142 +24,105 @@ def clean_test_case_output(text):
|
|
26 |
def generate_testcases(user_story):
|
27 |
"""
|
28 |
Generates advanced QA test cases based on a provided user story by interacting
|
29 |
-
with Nvidia's llama model API.
|
|
|
30 |
|
31 |
:param user_story: A string representing the user story for which to generate test cases.
|
32 |
-
:return: A list of
|
33 |
"""
|
|
|
|
|
34 |
few_shot_examples = """
|
|
|
35 |
"if its not a DropBury or ODAC Portal User Story, then we perform testing in Tech360 iOS App"
|
36 |
"Generate as many as testcases possible minimum 6 ,maximum it can be anything"
|
37 |
"Understand the story thoroughly"
|
38 |
"If it's a DropBury or ODAC Portal User Story, then we perform testing in ODAC Portal"
|
39 |
-
|
40 |
-
Please generate test cases in the following format:
|
41 |
-
|
42 |
-
Test Case 1:
|
43 |
-
Preconditions: [Describe any preconditions here]
|
44 |
-
Steps: [List the steps required to perform the test]
|
45 |
-
Expected Result: [Describe the expected result of the test]
|
46 |
-
|
47 |
-
Test Case 2:
|
48 |
-
Preconditions: [Describe any preconditions here]
|
49 |
-
Steps: [List the steps required to perform the test]
|
50 |
-
Expected Result: [Describe the expected result of the test]
|
51 |
"""
|
52 |
|
53 |
-
|
|
|
54 |
|
55 |
try:
|
|
|
56 |
completion = client.chat.completions.create(
|
57 |
-
model="meta/llama-3.1-405b-instruct",
|
58 |
messages=[
|
59 |
{"role": "user", "content": prompt}
|
60 |
],
|
61 |
-
temperature=0.
|
62 |
-
top_p=0.
|
63 |
-
max_tokens=4096,
|
64 |
-
stream=True
|
65 |
)
|
66 |
|
67 |
# Initialize an empty string to accumulate the response
|
68 |
test_cases_text = ""
|
69 |
-
|
70 |
# Accumulate the response from the streaming chunks
|
71 |
for chunk in completion:
|
72 |
if chunk.choices[0].delta.content is not None:
|
73 |
test_cases_text += chunk.choices[0].delta.content
|
74 |
|
75 |
-
# Print raw response for debugging
|
76 |
-
print("Raw response from model:", test_cases_text)
|
77 |
|
78 |
# Ensure the entire response is captured before cleaning
|
79 |
if test_cases_text.strip() == "":
|
80 |
-
return [{"
|
81 |
-
|
82 |
# Clean the output by unescaping HTML entities and replacing <br> tags
|
83 |
test_cases_text = clean_test_case_output(test_cases_text)
|
84 |
|
85 |
-
|
86 |
-
|
87 |
-
|
88 |
-
|
89 |
-
|
90 |
-
test_cases = []
|
91 |
-
for block in test_case_blocks:
|
92 |
-
lines = block.split('\n')
|
93 |
-
if len(lines) >= 4:
|
94 |
-
test_case = {
|
95 |
-
'Test Case': lines[0].replace('Test Case ', '').strip(),
|
96 |
-
'Preconditions': lines[1].replace('Preconditions: ', '').strip(),
|
97 |
-
'Steps': lines[2].replace('Steps: ', '').strip(),
|
98 |
-
'Expected Result': lines[3].replace('Expected Result: ', '').strip(),
|
99 |
-
}
|
100 |
-
test_cases.append(test_case)
|
101 |
-
|
102 |
-
if not test_cases:
|
103 |
-
return [{"Test Case": "No test cases generated or output was empty."}]
|
104 |
|
105 |
-
|
|
|
106 |
|
|
|
|
|
|
|
|
|
107 |
except requests.exceptions.RequestException as e:
|
108 |
print(f"API request failed: {str(e)}")
|
109 |
-
return [
|
110 |
|
111 |
-
|
112 |
-
|
113 |
-
Exports the test cases to an Excel file with specific columns:
|
114 |
-
- Test Case
|
115 |
-
- Preconditions
|
116 |
-
- Steps
|
117 |
-
- Expected Result
|
118 |
-
|
119 |
-
:param test_cases: A list of test case dictionaries.
|
120 |
-
:return: Bytes of the Excel file.
|
121 |
-
"""
|
122 |
if not test_cases:
|
123 |
return "No test cases to export."
|
124 |
|
125 |
-
|
126 |
-
|
127 |
-
|
128 |
-
|
129 |
-
|
130 |
-
|
131 |
-
|
132 |
-
|
133 |
-
|
134 |
-
|
135 |
-
|
136 |
-
|
137 |
-
|
138 |
-
|
139 |
-
|
140 |
-
|
141 |
-
|
142 |
-
|
143 |
-
|
144 |
-
|
145 |
-
|
146 |
-
|
147 |
-
|
148 |
-
|
149 |
-
|
150 |
-
|
151 |
-
|
152 |
-
|
153 |
-
|
154 |
-
|
155 |
-
|
156 |
-
|
157 |
-
# Adjust column widths for neatness
|
158 |
-
ws.column_dimensions['A'].width = 50 # Test Case
|
159 |
-
ws.column_dimensions['B'].width = 30 # Preconditions
|
160 |
-
ws.column_dimensions['C'].width = 50 # Steps
|
161 |
-
ws.column_dimensions['D'].width = 50 # Expected Result
|
162 |
-
|
163 |
-
output = io.BytesIO()
|
164 |
-
wb.save(output)
|
165 |
-
output.seek(0)
|
166 |
-
|
167 |
-
return output.getvalue()
|
|
|
1 |
import json
|
2 |
+
import csv
|
3 |
import io
|
4 |
import requests
|
5 |
import html # For escaping HTML characters
|
6 |
from bs4 import BeautifulSoup
|
7 |
+
from openai import OpenAI
|
|
|
|
|
|
|
8 |
|
9 |
# Initialize OpenAI API with Nvidia's Llama model
|
10 |
client = OpenAI(
|
|
|
24 |
def generate_testcases(user_story):
|
25 |
"""
|
26 |
Generates advanced QA test cases based on a provided user story by interacting
|
27 |
+
with Nvidia's llama model API. The prompt is refined for clarity,
|
28 |
+
and the output is processed for better quality.
|
29 |
|
30 |
:param user_story: A string representing the user story for which to generate test cases.
|
31 |
+
:return: A list of test cases in the form of dictionaries.
|
32 |
"""
|
33 |
+
|
34 |
+
# Few-shot learning examples to guide the model
|
35 |
few_shot_examples = """
|
36 |
+
|
37 |
"if its not a DropBury or ODAC Portal User Story, then we perform testing in Tech360 iOS App"
|
38 |
"Generate as many as testcases possible minimum 6 ,maximum it can be anything"
|
39 |
"Understand the story thoroughly"
|
40 |
"If it's a DropBury or ODAC Portal User Story, then we perform testing in ODAC Portal"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
41 |
"""
|
42 |
|
43 |
+
# Combine the few-shot examples with the user story for the model to process
|
44 |
+
prompt = few_shot_examples + f"\nUser Story: {user_story}\n"
|
45 |
|
46 |
try:
|
47 |
+
# Call the Nvidia llama API with the refined prompt
|
48 |
completion = client.chat.completions.create(
|
49 |
+
model="meta/llama-3.1-405b-instruct", # Using llama3.1 405b model
|
50 |
messages=[
|
51 |
{"role": "user", "content": prompt}
|
52 |
],
|
53 |
+
temperature=0.03, # Further lowering temperature for precise and deterministic output
|
54 |
+
top_p=0.7, # Prioritize high-probability tokens even more
|
55 |
+
max_tokens=4096, # Increase max tokens to allow longer content
|
56 |
+
stream=True # Streaming the response for faster retrieval
|
57 |
)
|
58 |
|
59 |
# Initialize an empty string to accumulate the response
|
60 |
test_cases_text = ""
|
61 |
+
|
62 |
# Accumulate the response from the streaming chunks
|
63 |
for chunk in completion:
|
64 |
if chunk.choices[0].delta.content is not None:
|
65 |
test_cases_text += chunk.choices[0].delta.content
|
66 |
|
|
|
|
|
67 |
|
68 |
# Ensure the entire response is captured before cleaning
|
69 |
if test_cases_text.strip() == "":
|
70 |
+
return [{"test_case": "No test cases generated or output was empty."}]
|
71 |
+
|
72 |
# Clean the output by unescaping HTML entities and replacing <br> tags
|
73 |
test_cases_text = clean_test_case_output(test_cases_text)
|
74 |
|
75 |
+
try:
|
76 |
+
# Try to parse the output as JSON, assuming the model returns structured test cases
|
77 |
+
test_cases = json.loads(test_cases_text)
|
78 |
+
if isinstance(test_cases, list):
|
79 |
+
return test_cases # Return structured test cases
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
80 |
|
81 |
+
else:
|
82 |
+
return [{"test_case": test_cases_text}] # Return as a list with the text wrapped in a dict
|
83 |
|
84 |
+
except json.JSONDecodeError:
|
85 |
+
# Fallback: return the raw text if JSON parsing fails
|
86 |
+
return [{"test_case": test_cases_text}]
|
87 |
+
|
88 |
except requests.exceptions.RequestException as e:
|
89 |
print(f"API request failed: {str(e)}")
|
90 |
+
return []
|
91 |
|
92 |
+
# Add options for multiple test case formats
|
93 |
+
def export_test_cases(test_cases, format='json'):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
94 |
if not test_cases:
|
95 |
return "No test cases to export."
|
96 |
|
97 |
+
# Convert test cases (which are currently strings) into a structured format for CSV
|
98 |
+
structured_test_cases = [{'Test Case': case} for case in test_cases]
|
99 |
+
|
100 |
+
if format == 'json':
|
101 |
+
# Improve JSON export to be line-by-line formatted
|
102 |
+
return json.dumps(test_cases, indent=4, separators=(',', ': ')) # More readable format
|
103 |
+
elif format == 'csv':
|
104 |
+
if isinstance(test_cases, list) and isinstance(test_cases[0], dict):
|
105 |
+
output = io.StringIO()
|
106 |
+
csv_writer = csv.DictWriter(output, fieldnames=test_cases[0].keys(), quoting=csv.QUOTE_ALL)
|
107 |
+
csv_writer.writeheader()
|
108 |
+
csv_writer.writerows(test_cases)
|
109 |
+
return output.getvalue()
|
110 |
+
else:
|
111 |
+
raise ValueError("Test cases must be a list of dictionaries for CSV export.")
|
112 |
+
|
113 |
+
# 2. Save test cases as a downloadable file
|
114 |
+
def save_test_cases_as_file(test_cases, format='json'):
|
115 |
+
if not test_cases:
|
116 |
+
return "No test cases to save."
|
117 |
+
|
118 |
+
if format == 'json':
|
119 |
+
with open('test_cases.json', 'w') as f:
|
120 |
+
json.dump(test_cases, f)
|
121 |
+
elif format == 'csv':
|
122 |
+
with open('test_cases.csv', 'w', newline='') as file:
|
123 |
+
dict_writer = csv.DictWriter(file, fieldnames=test_cases[0].keys())
|
124 |
+
dict_writer.writeheader()
|
125 |
+
dict_writer.writerows(test_cases)
|
126 |
+
else:
|
127 |
+
return f"Unsupported format: {format}"
|
128 |
+
return f'{format} file saved'
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|